A GC-MS untargeted metabolomics approach for the classification of chemical differences in grape juices based on fungal pathogen.

Fungal bunch rot of grapes leads to production of detrimental flavour compounds, some of which are well characterised but others remain unidentified. The current study uses an untargeted metabolomics approach to classify volatile profiles of grape juices based on the presence of different fungal pathogens. Individual grape berries were inoculated with Botrytis cinerea, Penicillium expansum, Aspergillus niger or A. carbonarius. Grape bunches were inoculated and blended with healthy fruit, to provide 10% (w/w) infected juice. Juices from the above sample batches were analysed by GC/MS. PLS-DA of the normalised summed mass ions indicated sample classification according to pathogen. Compounds identified from those mass ion matrices that had high discriminative value for classification included 1,5-dimethylnaphthalene and several unidentified sesquiterpenes that were relatively higher in B. cinerea infected samples. A. niger and A. carbonarius samples were relatively higher in 2-(4-hexyl-2,5-dioxo-2,5-dihydrofuran-3-yl)acetic acid, while P. expansum samples were higher in γ-nonalactone and m-cresol.